Note: This Paper Was Reviewed and Rejected (Ms #08-0125)

Anesthesia for Earthquake Victims in Wenchuan County Seat

Qingxiang Mao, MD1, Hong Yan, MD,1 Quan Du, MD,1 Hengjiang Ge, MD,1 Jianrong Zeng, MD,2 Jihong Zhou, MD3

1. Department of Anesthesiology of Daping Hospital, Institute of Surgery, Third Military Medical University, Chongqing, China.

2. Wenchuan People’s Hospital, Aba Prefecture, Sichuan Province, China.

3. Fourth Unit, Institute of Surgery, Daping Hospital, Third Military Medical University, Chongqing, China.

Corresponding author: Hengjiang Ge, MD, Professor, Department of Anesthesiology of Daping Hospital Institute of Surgery Third Military Medical University Chongqing, 400042, China Tel: 86-23-68757081 E-mail: [email protected]

Short Title: Anesthesia for earthquake-affected victims

Support was provided solely from departmental sources. 2

ABSTRACT

Background: On May 12, 2008, a massive earthquake shook the Wenchuan district of China. A medical team from Daping Hospital, Third Military Medical University was sent to Wenchuan county seat to provide medical relief. We here characterize the perioperative profiles of the earthquake victims and share the experiences of this mission.

Methods: The clinical data of the 151 patients admitted from May 12 to May 20 were collected.

Profiles of trauma, surgeries, anesthesia, drugs and anesthesia-related complications were analyzed. Student t-test, one-way ANOVA and Cox proportional-hazards analysis were performed.

Results: Among 151 admitted patients, 143 were trauma patients, 8 were non-earthquake affected patients, and 14 cases died after admission. Injury Severity Score (ISS), intracranial injury, thoracic injury and pelvic fracture were independent prognostic factors to mortality.

Sixty-eight operations were performed. Most of them were debridement for fractures of the extremities. Regional and neuraxial anesthesia were the major methods of anesthesia. The ISS of operative patients was higher than that of non-operative patients. However, there was no significant correlation between ISS and fluid dosage or duration of surgery. Most anesthesia complications were those common in daily clinical practice, except for tension pneumothorax and aggravation of posttraumatic stress disorder.

Conclusions: Ninety-six hours after the earthquake, most surgery patients had injuries of the extremities. The most common type of anesthesia provided was neuraxial anesthesia and nerve blocks. 3

INTRODUCTION

On May 12, 2008, a strong earthquake measuring 8.0 on the Richter scale shook the Wenchuan district of China. According to the statistic data announced by the Central People’s Government on July 29, 2008, 69,200 people were killed. An additional 374,216 people were injured, and

18,195 people were listed as missing. Destroyed infrastructure, bad weather, a mountainous terrain, and scattered habitations created difficulties for rescue personnel. Nonetheless, hundreds of medical teams, including our team (one of several dispatched from Daping Hospital, Third

Military Medical University) were sent to the disaster area.

We departed 22 hours after the earthquake and arrived in Dujiangyan in 7 hours. Although

Wenchuan is only a 6-hour drive from Dujiangyan, the road to Wenchuan was totally destroyed, requiring us to make a detour of 1000 kilometers. Continually hampered by aftershocks -- one minibus of ours was crushed by a landslide following an aftershock of 5.9 -- we finally arrived at the Wenchuan county seat, on May 16, 96 hours after the earthquake struck. Our medical relief team was the first to arrive on the scene.

Our objective in this article is to characterize the trauma to emergency surgical patients and the perioperative profiles of the earthquake victims. Furthermore, we offer recommendations to assist future disaster relief efforts in isolated, rural regions. 4

MATERIALS AND METHODS

After the earthquake, the faculty of local hospitals immediately activated medical relief efforts.

Wenchuan People’s Hospital, the facility in the area, and other smaller hospitals were merged into a 50-bed tent hospital before our arrival and treated a large number of casualties. Because of a lack of specialists, and because it had lost most of its equipment, this tent hospital only performed debridement, abdominal delivery, appendectomy, and splenectomy. After our arrival, our medical team joined with those providing care with the tent hospital. Our team was in charge of surgery and the triage of patients. Starting on May 21, most patients in the tent hospital were evacuated to outside hospitals by helicopter.

We retrospectively analyzed the medical records of patients admitted to the tent hospital from

May 12 to May 20. Patients not admitted were excluded from the study. Demographic characteristics and surgical and anesthesia profiles were included in our analysis.

All data were processed and analyzed by SPSS13.0 for Windows (SPSS Inc., Chicago, IL). The data for Injury Severity Score (ISS), age, fluid dosage and duration of surgery were presented as means ±SD (standard deviation). Age distribution was analyzed using a SPSS chart tool. Risk factors for mortality in victims were tested by Cox proportional-hazards analysis. A significant difference was defined as a P value < 0.05. 5

RESULTS

Mission preparation and logistical support

Our medical team comprised 6 orthopedists, 3 general surgeons, 2 urologists, 2 thoracic surgeons, 2 neurosurgeons, 3 anesthesiologists, 1 ophthalmologist, 1 cardiologist, 1 respiratory physician, 1 laboratory technician, 1 ultrasound technician, and 4 nurses (including 3 operating room nurses). Each specialist held a post-graduate or doctorate degree in his or her specialty. As the anesthesia support needed for this rescue was unknown, we brought drugs, fluids, equipment, and other anesthesia accessories sufficient for 200 cases of traumatic surgeries (Table 1).

Nonetheless, there were deficiencies in our preparation for this mission. For example, we did not bring blood products because we did not have the necessary transport vehicle. The same was true for radiological apparatus and some special drugs (Table 1).

In addition, logistical support was not well planned. Emergency generators, water purifiers, and satellite communication equipment were not part of our preparation inventory. These deficiencies negatively affected our performance. Although visceral leishmaniasis and anthracnose were abundant in Wenchuan district, no measures were taken to protect team members from these and other contagious diseases.

Triage and admissions

The biggest local hospital, Wenchuan People’s Hospital, began to treat the injured victims within hours after the earthquake. Most of the patients were sent by family members or neighbors. Late in the afternoon, several other hospitals were merged into Wenchuan People’s Hospital to carry 6 out medical relief more effectively. Triage was performed by 1 surgeon, 1 internist, and 2 nurses.

Severely injured or infected and immobile patients were hospitalized for review. Emergency surgeries were performed when possible. The operation tent contained only one old anesthesia machine, one monitor, and several tables. Patients with soft tissue or skin contusion were treated by another group of doctors and cared for by family members. Food, water, and tents were provided by the local government.

This tent hospital was erected on higher ground to avoid flooding from the local reservoir. After we arrived, we were placed in charge of triage and surgical services. All hospitalized patients were reevaluated. Patient treatment came under the supervision of our specialists. When possible, evacuation to outside hospitals was arranged.

As of May 20, 8 days after the disaster, 816 trauma patients were triaged and 151 cases (84 males, 67 females) were admitted to the tent hospital, including 8 non-earthquake affected patients (delivery and appendicitis, Table 2). Age distribution of admitted victims is shown in

Figure 1. Seventy-two hours after the earthquake, 137 cases had been hospitalized. Fourteen patients died after admission.

Trauma profiles

Among 816 patients, 6 cases (0.74%) had intracranial injury, 8 cases (0.98%) had thoracic injury, 5 cases (0.61%) had abdominal injury (1 case of splenic rupture and 4 cases of renal contusion), 10 cases (1.23%) had spine fractures, 8 cases (0.98%) had pelvic fractures, 6 cases

(0.74%) had cephalofacial injury, 19 cases (2.33%) had upper limb fracture, 54 cases (6.62%) 7 had lower limb fracture, 27 cases (3.31%)had multi-site injuries and 673 cases (82.48%) had superficial lacerations or soft tissue sprain.

One hundred forty-three admitted patients (excluding 8 non-earthquake affected patients) were evaluated using the ISS (from “Trauma Database of Chinese Society for Trauma,” copyright by

Institute of Traffic Medicine, Daping Hospital, Third Military Medical University, Chongqing,

China). Patients admitted on May 12 had the highest ISS scores (Table 3), and the most severely injured of these patients were quickly sent to the local hospital.

Fourteen patients died at the tent hospital from May 12 to May 20. The causes of death were hemorrhagic shock (6 cases), crush syndrome (3 cases), intracranial hematoma (2 cases), hemopneumothorax (1 case), and acute renal failure (2 cases). The ISS scores of those patients who died were obtained from medical records. Thirteen potential factors including gender, age, time from injury to admission, ISS, intracranial injury, thoracic injury, spine fracture, abdominal injury, pelvic fracture, cephalofacial injury, upper limb fracture, lower limb fracture, and infection were analyzed using a Cox proportional hazards model. Only ISS, intracranial injury, thoracic injury and pelvic fracture were independent risk factors for mortality (Table 4).

Surgical patient condition and surgical profiles

Because of limited medical resources, the priority of emergency surgery was decided by importance and severity of the injured organs. During this medical relief, surgical priorities in sequence were life-threatening brain trauma, chest or abdomen injury, spine fracture with paraplegia, trauma with severe infection, abdominal delivery, open extremity fracture, 8 appendectomy, and closed extremity fracture. There were no brain or chest surgeries performed during this mission.

There were 42 male and 23 female admitted patients undergoing surgeries, and ASA physical status classifications of 1~2 and 3 were present in 63 and 2 cases, respectively. Beginning on

May 21, most of the admitted patients were evacuated to near hospitals. Of 65 surgery patients, there were 55 cases (84%) of extremity fractures, 5 cases of appendicitis (8%), 3 cases of abdominal delivery (5%), 1 case of abdominal injury, and 1 case of spine fracture. One patient received surgery twice and 1 patient was operated on three times, for a total of 68 surgeries.

Of these 68 operations, 50 (74%) were for debridement of fracture or infection, 5 (7%) were for amputation, 5 (7%) were appendectomies, 3 were (4%) internal fixations of closed fracture, 3 operations were cesarean deliveries (4%), 1 was for internal fixation of spine fracture and 1

(1.5%) was a splenectomy. As shown in Table 5, there was no difference between non-operated and operated patients in regard to age. As expected, ISS scores of operated patients were higher than those of non-operation patients.

Anesthesia profiles

In the local hospital, general anesthesia, epidural anesthesia, and brachial plexus nerve block were performed by 2 anesthesiologists. The surgeons also performed epidural anesthesia. During our medical relief, regional and neuraxial anesthesia (epidural and spinal) were preferred in patients with isolated injuries without contraindication (1). There were 36 cases (53%) of epidural anesthesia, 15 cases (22%) of spinal-epidural anesthesia, 8 cases (11%) of total IV 9 anesthesia (TIVA) with endotracheal intubation, 5 cases (7.4%) of brachial plexus nerve block, 3 cases (4%) of spinal anesthesia, and 1 case of TIVA with mask ventilation. The average duration of anesthesia was 68 ± 29 min (range: 15 - 165 min). We had neither anesthesia machines nor ventilators, so we were unable to use inhaled anesthetics. There was no perioperative mortality during our relief mission.

We carried enough crystalloid (saline, lactated Ringer's solution, and acetated Ringer’s solution) and colloid (succinylated gelatin and hydroxyethyl starch) solutions to provide fluid resuscitation. Blood products were provided by the local hospitals. Every patient received crystalloid solutions, with the mean (SD) volume of 786 ± 377 mls. Thirty-eight patients also received colloid solutions, with a mean (SD) volume of 692 ± 257 mls. Eight patients received red blood cell, with a mean (SD) volume of 650 ± 207 mls. One patient received 800 mls of plasma.

Drugs used in this mission are shown in Table 6. Lidocaine was the drug used most often.

Epinephrine was mainly used in a mixture with lidocaine (1:200 000). Dexamethasone was used to prevent transfusion reaction and asthmatic attack.

Perioperative anesthesia complications

Anesthesia-related complications are shown in Table 7. The incidence of hypotension (systolic blood pressure < 90 mmHg or < 70% of baseline) after neuraxial anesthesia was 38%. Nausea and vomiting during neuraxial anesthesia were mainly caused by hypotension or unopposed vagal tone during neuraxial anesthesia, as judged by time of onset and its response to ephedrine. 10

Five patients had from incomplete epidural block and were managed by combining IV anesthesia under mask, first with an IV fentanyl and droperidol, followed by IV midazolam and sulfentanil.

Continuous infusion of propofol was used as a last resort. One patient with postoperative hypoxemia was obese and obstructive sleep apnea was suspected. The patient who had dural puncture was treated with epidural infusion of gelatin to prevent headache. Tension pneumothorax occurred in a patient with multiple penetrating injuries undergoing general anesthesia, who presented with high airway pressure, decreasing SpO2, and hemodynamic instability. After closed drainage was performed, the patient’s condition improved.

Before our arrival, the local hospital staff performed several debridement surgeries under local anesthesia. One female patient suffered too much pain during operation. Several days after this operation she was increasingly unstable and refused surgery. This indicated that unsuccessful anesthesia could aggravate the psychic trauma of earthquake victims. 11

DISCUSSION

Major earthquakes can produce overwhelming numbers of casualties (2). Medical assistance from unaffected areas is crucial and must be immediate if they are to be effective. This study aimed to share the experience of our medical team treating patients admitted to the only hospital in the Wenchuan county seat after the May 12 earthquake.

Although we spent an entire day planning our mission before leaving for the affected area, we did not adequately prepare for every contingency. In retrospect, we conclude that:

1. a medical team should be self-sustaining and provided with basic facilities and

equipment, including temporary wards, an operating room, emergency generators, water

purification equipment, communication equipment, transportation, food, and lodging;

2. materials should be sufficient to cover not only the needs of the basic first-responder

operations but must also include diagnostic capabilities and intensive care supplies;

3. staff must be provided with vaccinations for specific contagious diseases, ideally given

many months in anticipation of future medical rescue missions, and

4. staff should be regularly trained on emergency rescue skills in anticipation of future

medical rescue missions.

The county seat of Wenchuan, Weizhou Town, with a population of 30,000, had fewer casualties than Yingxiu Town, another precinct of Wenchuan, with a population of 16,000. Weizhou Town lost fewer than 500 people while Yingxiu Town lost more than 5000 in the earthquake. This may 12 explain some difference in the extent of traumatic injury in this county seat (3) where only a few patients were admitted.

According our data, 14 patients admitted to the tent hospital died before we arrived, and most of them were sent to the hospital within 24 hours of the earthquake. Rescue time has a decisive effect on survival rate (2). Interestingly, time from injury to admission was not associated with mortality in our Cox regression analysis while ISS, intracranial, thoracic injury and pelvic fracture were independent predictors of mortality. This phenomenon may be explained by the absence of neurosurgeons, thoracic surgeons, or other specialists in the local hospital, as well as the 96 hour delay from the earthquake to our arrival.

According to our data, most of our surgeries were debridement of extremities, especially within the first 96 hours after the earthquake. Regional and neuraxial anesthesia was our primary choice because of several obvious advantages:

1. they are easy to perform;

2. the reduce the risk of aspiration;

3. they reduce the risk of managing a difficult airway;

4. they require less postoperative care;

5. they provide better postoperative analgesia;

6. they do not impede bowel function; and

7. they do not require as much supplementary oxygen (4,5). 13

The ISS is based on the Abbreviated Injury Scale severity values derived from the sum of the squares of the highest Abbreviated Injury Scale values from each of the three most severely injured body regions (7). Despite its use, the ISS has been criticized because of its limitations

(8). We studied the association of the ISS with fluid dosage, urgency and duration of surgeries, and did not find correlations. However, we did find that ISS predicted the urgency of surgery

(Table 5). These results are similar to those of other studies. (8).

The anesthesia-related complications we encountered in this medical relief effort were those common in daily clinical practice. However, there were two complications that deserve mention.

Tension pneumothorax can be rapidly fatal, but is readily identified through clinical and radiological evaluations. Tension pneumothorax should be always suspected in the presence of chest trauma before mechanical ventilation, especially in field conditions. Most patients in the tent hospital showed evidence of post traumatic stress disorder, which was exacerbated in one woman after undergoing surgery with unsuccessful anesthesia. Some studies show that pain, and some anesthetics like ketamine, can exacerbate posttraumatic stress disorder (9, 10). During the perioperative period adequate anesthesia should be provided. It might also be wise to avoid using ketamine. In addition, postoperative analgesia should be provided. 14

REFERENCES:

1. Brown DL. Spinal, epidural, and caudal anesthesia. In: Miller RD, eds. Miller’s anesthesia, sixth edition. Philadelphia: Elsevier/Churchill Livingstone, 2005: 1654-55.

2. Wang ZG, Song SM. Medical aid in earthquake disasters. Chin J Traumatol 2008; 11: 195-7.

3. Mulvey JM, Awan SU, Qadri AA, Maqsood MA. Profile of injuries arising from the 2005

Kashmir Earthquake: The first 72 h. Injury 2008;39:554-60.

4. Liu SS, Strodtbeck WM, Richman JM, Wu CL. A comparison of regional versus general anesthesia for ambulatory anesthesia: a meta-analysis of randomized controlled trials. Anesth

Analg 2005;101:1634-42.

5. Steinbrook RA. Epidural anesthesia and gastrointestinal motility. Anesth Analg 1998; 86:837-

6. Charuluxananan S, Bunburaphong P, Tuchinda L, Vorapaluk P, Kyokong O. Anesthesia for

Indian Ocean tsunami-affected patients at a southern Thailand provincial hospital. Acta-

Anaesthesiol-Scand 2006;50: 320-3.

7. Linn S. The Injury Severity Score-Importance and Uses. Ann-Epidemiol 1995; 5: 440-6.

8. Baxt WG, Upenieks V. The lack of full correlation between the Injury Severity Score and the resource needs of injured patients. Ann-Emerg-Med. 1990; 19: 1396-400.

9. Zatzick DF, Rivara FP, Nathens AB, Jurkovich GJ, Wang J, Fan MY, Russo J, Salkever DS,

Mackenzie EJ. A nationwide US study of post-traumatic stress after hospitalization for physical injury. Psychol-Med 2007; 37: 1469-80.

10. Schonenberg M, Reichwald U, Domes G, Badke A, Hautzinger M. Effects of peritraumatic ketamine medication on early and sustained posttraumatic stress symptoms in moderately injured accident victims. Psychopharmacology-(Berl) 2005; 182: 420-5. 15

Table 1. Materials initially prepared and others still needed during this mission

Categorization Planned in this mission Still needed Drugs Local anesthetics Inhalation anesthetics IV anesthetics (non-opioid) Insulin Narcotic analgesic Oxytocin Cardiovascular drugs Muscle relaxants Asthma drugs Coagulant and anticoagulant agents corticosteroids and antiallergic agents Diuretics Fluids and Blood Crystalloid fluid Erythrocyte suspension Colloid fluid Fresh frozen plasma products

Cryoprecipitate Platelet Equipment Portable ventilators Anesthesia machines Monitors Imaging diagnostic devices Defibrillator Laboratory examination devices Portable blood gas analyzer Oxygenator Ultrasonic diagnostic apparatus Anesthesia Epidural/Spinal puncture pack Intubation Accessories Accessories Central venous/arterial catheter

Table 2: Demographic data for patients admitted each day after the earthquake

Non-earthquake Cases / Date Male / Female Age (years) Died Total Cases May 12 1/63 41/22 40 ± 16.5 10 May 13 0/24 11/13 39 ± 20.3 2 May 14 1/16 10/6 34 ± 13.5 1 May 15 2/26 11/15 40 ± 20.5 1 May 16 2/7 4/3 56 ± 15.0 0 May 17 1/4 2/2 43 ± 34.6 0 May 18 1/5 2/3 37 ± 13.3 0 May 19 0/4 3/1 33 ± 15.5 0 May 20 0/2 0/2 37 ± 2.9 0 17

Table 3: Injury severity score (ISS) of the admitted trauma patients

Date < 9 9 – 15 > 15 Average ISS May 12 34 15 13 9.1 ± 6.90 May 13 15 8 1 6.1 ± 5.44 May 14 7 8 0 7.0 ± 3.02 May 15 17 7 0 5.4 ± 3.23 May 16 5 0 0 1.6 ± 0.55 May 17 1 2 0 6.3 ± 4.62 May 18 3 1 0 3.8 ± 3.77 May 19 3 1 0 5.5 ± 2.38 May 20 2 0 0 5.0 ± 4.24 18

Table 4: Results of multivariable Cox regression analysis, including gender, age, time from injury to admission, ISS, intracranial injury, thoracic injury, spine fracture, abdominal injury, pelvic fracture, cephalofacial injury, upper limb fracture, lower limb fracture and infection. Only independent significant predictors are shown.

Variable RR 95% CI P value ISS 1.2 1.1 - 1.3 0.000 Thoracic injury 7.0 1.5 – 32 0.013 Pelvic fracture 11.8 2.6 – 52 0.001 Intracranial injury 9.7 1.4 - 67 0.023

RR: relative risk, CI: confidence interval; ISS: injury severity score. 19

Table 5: Age and ISS in non-surgical and surgical patients

Variable Non-surgical patients (n=72) Surgical patients (n=57*) Age (years) 39 ± 17.9 39 ± 17.6 ISS 4.6 ± 2.8 7.8 ± 3.2

* 8 non-earthquake affected patients were excluded. Patients who underwent operation two or three times were only counted once in this analysis. ISS: injury severity score. 20

Table 6: Perioperative consumption of different drugs

Generic names Total number of ampules Number of cases 2% Lidocaine（100mg in 5ml ampule） 137 39 Ephedrine（30mg in 1ml ampule） 24 24 Propofol（200mg in 20ml ampule） 21 12 Fentanyl（0.2mg in 2ml ampule） 19 12 Epinephrine（1mg in 1ml ampule） 19 19 Vecuronium（4mg per ampule） 18 8 0.75% Bupivacaine（37.5mg in 5ml ampule） 18 18 0.75% Ropivacaine（75mg in 10ml ampule） 13 13 Midazolam（10mg in 2ml ampule） 9 9 Dexamethasone（5mg in 1ml ampule） 8 8 Droperidol（5mg in 2ml ampule） 5 5 Sufentanil（50μg in 1ml ampule） 3 3 Etomidate（20mg in 10ml ampule） 1 1 21

Table 7: Perioperative anesthetic complications

------Neuraxial anesthesia ------General anesthesia Incomplete Nausea and Dural Postoperative Tension Variable Hypotension epidural vomiting puncture hypoxemia pneumothorax block Number of 24 5 2 1 1 1 cases Percentage (%) 35 7.4 2.9 1.5 1.5 1.5 (n=68) 22

Figure 1: Age distribution of earthquake victims admitted to the tent hospital in Wenchuan